Re: Documenting GridFitTT

On Mon, Oct 27, 2003 at 04:52:28PM +0300, Igor V. Melichev wrote:

> Attaching an improved edition.

Thanks. Language and Use patches are okay to commit. I'll just assume 
you got all the new files in the Develop.htm patch. BTW, technically, 
they files in each section are separated by commas, with a period at the 
end of the list, but I don't think that's so important. I approve commit 
either way.

> > Presumedly the rendering differences are caused by the
> > former, since the later is off by default?
> 
> No, it is not.
> 
> The old code for a TT oulline decoding is buggy -
> it accumulates an arithmetic error while scaling the outline
> to device resolution. The new code fixed this error.

Ok, so this is the origin of the rendering differences, and the default 
is still GridFitTT=0?

> Uuhhh, Graham Asher's work is fully another story.
> It implements a bridge to FreeType 2. In this case an user needs to
> download 2 projects (GS and full FreeType 2), build them,
> prepare special configuration files in gs/lib and run then.
> It's a monster.

This was my point. Your patch to Projects.htm *replaces* the entry 
dealing with freetype integration with one describing an improvement to 
the ghostscript TT interpreter, but they are separate projects. I think 
we should keep the old entry as well, amended to reflect the reduced 
priority due to your improvements to the GS rasterizer.

> > My understanding from Raph was that there was little merit to this. Can
> > you clarify what you mean, just for my own information?
> 
> Until now I'm unclear about Raph's position with it. Sometimes it says
> white, sometimes black. Maybe depending on whether.
> Ray strongly wants to "port" FreeType, but I believe that his wish
> completely ignores the real resource (Igor's skill), assigned for doing
> that.
> I put this phrase into Projects.htm just for non-teasing (those 2) gooses.

Hmm. I don't think that idiom translates.

 -r